vantage of battery backed write cache...
Regards,
Mikael
-Original Message-
From: Guoping Zhang [mailto:[EMAIL PROTECTED]
Sent: den 28 april 2006 07:35
To: Mikael Carneholm; pgsql-performance@postgresql.org
Cc: Guoping Zhang (E-mail)
Subject: RE: [PERFORM] how unsafe (or worst scena
Message-
From: Tom Lane [mailto:[EMAIL PROTECTED]
Sent: 2006Äê4ÔÂ28ÈÕ 14:57
To: [EMAIL PROTECTED]
Cc: pgsql-performance@postgresql.org; 'Guoping Zhang (E-mail)'
Subject: Re: [PERFORM] how unsafe (or worst scenarios) when setting
fsync OFF for postgresql
"Guoping Zhang&quo
erformance@postgresql.org; Guoping Zhang (E-mail)
Subject: Re: [PERFORM] how unsafe (or worst scenarios) when setting
fsync OFF for postgresql
Guoping,
On 4/27/06, Guoping Zhang <[EMAIL PROTECTED]> wrote:
> We have to looking at setting fsync OFF option for performance reason,
Did you try
"Guoping Zhang" <[EMAIL PROTECTED]> writes:
> a) The tests consists of ten thousands very small transactions, which are
> not grouped, that is why so slow with compare to set fsync off.
Yup.
> c) wal_sync_method is set to 'open_datasync', which is fastest among the
> four, right?
Well, is it? Y
Lane
Sent: 2006Äê4ÔÂ28ÈÕ 0:53
To: [EMAIL PROTECTED]
Cc: pgsql-performance@postgresql.org; Guoping Zhang (E-mail)
Subject: Re: [PERFORM] how unsafe (or worst scenarios) when setting
fsync OFF for postgresql
"Guoping Zhang" <[EMAIL PROTECTED]> writes:
> Our application has a st
"Guoping Zhang" <[EMAIL PROTECTED]> writes:
> Our application has a strict speed requirement for DB operation. Our tests
> show that it takes about 10secs for the operation when setting fsync off,
> but takes about 70 seconds when setting fsync ON (with other WAL related
> parametered tuned).
I ca
Guoping,
On 4/27/06, Guoping Zhang <[EMAIL PROTECTED]> wrote:
> We have to looking at setting fsync OFF option for performance reason,
Did you try the other wal sync methods (fdatasync in particular)? I
saw a few posts lately explaining how changing sync method can affect
performances in specific
Get a SCSI controller with a battery backed cache, and mount the disks
with data=writeback (if you use ext3). If you loose power in the middle
of a transaction, the battery will ensure that the write operation still
completes. With asynch writing setup like this, fsync operations will
return almost
